An algorithm for the automated generation of rheological models

This paper presents a computational approach to the generation of rheological models. A rheological structure can be analogically described as a set of basic components connected in series or in parallel. The models are automatically created in two different forms: a symbolic qualitative relation and a mathematical equation. In both cases, the designed algorithm uses the same knowledge representation scheme which is based on rooted binary tree-like graphs. The mathematical model of a rheological structure, made up of n basic components, is built from the basic models of each component by exploiting connection laws. This work is part of a more ambitious project aiming at carrying out a system for automated reasoning about rheological systems. Beside a model-building task, such a system should perform a simulation and diagnostic task. Therefore it should provide methods for the simulation, both qualitative and quantitative, of the behavior of a rheological structure and methods for the identification of a model of an actual material, given its behavior.